This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [RFC][06/19] Target FP: Use print_floating in tdep code
- From: "Maciej W. Rozycki" <macro at imgtec dot com>
- To: Ulrich Weigand <uweigand at de dot ibm dot com>
- Cc: <gdb-patches at sourceware dot org>
- Date: Wed, 4 Oct 2017 23:26:30 +0100
- Subject: Re: [RFC][06/19] Target FP: Use print_floating in tdep code
- Authentication-results: sourceware.org; auth=none
- References: <20170905182100.37B12D8086F@oc3748833570.ibm.com>
Hi Ulrich,
Thank you for your effort in getting this area right, the limitations in
our FP arithmetic handling really troubled me.
> In addition, both mips-tdep and sh64-tdep use host FP to print
> floating-point values. This patch changes them to print_floating as well.
> Again, this may change the output format a bit, but in general it would
> appear preferable to use the default print_floating now (this will also
> now match how all other FP values are printed).
I pushed your changes 01-06 combined through testing with a couple of
MIPS/Linux targets, native and crossed (from x86-64), and all have
consistently regressed in gdb.base/dfp-exprs.exp, e.g.:
(gdb) p 1.2df
-$1 = 1.2
-(gdb) PASS: gdb.base/dfp-exprs.exp: p 1.2df
+$1 =
+(gdb) FAIL: gdb.base/dfp-exprs.exp: p 1.2df
p -1.2df
-$2 = -1.2
-(gdb) PASS: gdb.base/dfp-exprs.exp: p -1.2df
+$2 =
+(gdb) FAIL: gdb.base/dfp-exprs.exp: p -1.2df
i.e. values to be output are missing, numerous across this test case. No
other regressions though.
Also I find the new formatting of `info all-registers' output
functionally regressed (rather than merely "changed a bit"), e.g.:
- f0: 0x00000000 flt: 0 dbl: 0
- f1: 0x00000000 flt: 0
- f2: 0xd2f1a9fc flt: -5.18969491e+11 dbl: -0.001
- f3: 0xbf50624d flt: -0.813999951
+ f0: 0x00000000 flt: 0 dbl: 0
+ f1: 0x00000000 flt: 0
+ f2: 0xd2f1a9fc flt: -5.18969491e+11 dbl: -0.001
+ f3: 0xbf50624d flt: -0.813999951
or taking a complete listing what used to look like:
(gdb) info all-registers
zero at v0 v1 a0 a1 a2 a3
R0 00000000 00000001 00000001 d2f1a9fc 00000000 00000000 00000fff 00000000
t0 t1 t2 t3 t4 t5 t6 t7
R8 00000000 00021000 00000000 00000417 00000000 8144f6e0 8122d6b8 00000000
s0 s1 s2 s3 s4 s5 s6 s7
R16 00000000 00000000 00000000 00500000 0052de28 0052e008 00000000 00000000
t8 t9 k0 k1 gp sp s8 ra
R24 77db4000 77e3f390 00000000 00000000 0041e220 7fff6090 7fff6090 00405c2c
status lo hi badvaddr cause pc
00009cf3 0046790f 00000005 00417004 00800024 00405c30
f0: 0x00000000 flt: 0 dbl: 0
f1: 0x00000000 flt: 0
f2: 0xd2f1a9fc flt: -5.18969491e+11 dbl: -0.001
f3: 0xbf50624d flt: -0.813999951
f4: 0x7ff80000 flt: nan dbl: nan
f5: 0x7ff80000 flt: nan
f6: 0x7ff80000 flt: nan dbl: nan
f7: 0x7ff80000 flt: nan
f8: 0x7ff80000 flt: nan dbl: nan
f9: 0x7ff80000 flt: nan
f10: 0x7ff80000 flt: nan dbl: nan
f11: 0x7ff80000 flt: nan
f12: 0x45a1cac1 flt: 5177.34424 dbl: 45.654000000000003
f13: 0x4046d3b6 flt: 3.10667181
f14: 0x70a3d70a flt: 4.05648183e+29 dbl: -67.659999999999997
f15: 0xc050ea3d flt: -3.26429677
f16: 0x7ff80000 flt: nan dbl: nan
f17: 0x7ff80000 flt: nan
f18: 0x7ff80000 flt: nan dbl: nan
f19: 0x7ff80000 flt: nan
f20: 0x7ff80000 flt: nan dbl: nan
f21: 0x7ff80000 flt: nan
f22: 0x7ff80000 flt: nan dbl: nan
f23: 0x7ff80000 flt: nan
f24: 0x7ff80000 flt: nan dbl: nan
f25: 0x7ff80000 flt: nan
f26: 0x7ff80000 flt: nan dbl: nan
f27: 0x7ff80000 flt: nan
f28: 0x7ff80000 flt: nan dbl: nan
f29: 0x7ff80000 flt: nan
f30: 0x7ff80000 flt: nan dbl: nan
f31: 0x7ff80000 flt: nan
fcsr fir restart
00800000 00f30000 00000000
(gdb)
is now:
(gdb) info all-registers
zero at v0 v1 a0 a1 a2 a3
R0 00000000 00000001 00000001 d2f1a9fc 00000000 00000000 00000fff 00000000
t0 t1 t2 t3 t4 t5 t6 t7
R8 00000000 00021000 00000000 00000417 00000000 8144f6e0 8122d6b8 00000000
s0 s1 s2 s3 s4 s5 s6 s7
R16 00000000 00000000 00000000 00500000 0052de28 0052e008 00000000 00000000
t8 t9 k0 k1 gp sp s8 ra
R24 77db4000 77e3f390 00000000 00000000 0041e220 7fff6090 7fff6090 00405c2c
status lo hi badvaddr cause pc
00009cf3 0046790f 00000005 00417004 00800024 00405c30
f0: 0x00000000 flt: 0 dbl: 0
f1: 0x00000000 flt: 0
f2: 0xd2f1a9fc flt: -5.18969491e+11 dbl: -0.001
f3: 0xbf50624d flt: -0.813999951
f4: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f5: 0x7ff80000 flt: nan(0x780000)
f6: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f7: 0x7ff80000 flt: nan(0x780000)
f8: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f9: 0x7ff80000 flt: nan(0x780000)
f10: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f11: 0x7ff80000 flt: nan(0x780000)
f12: 0x45a1cac1 flt: 5177.34424 dbl: 45.654000000000003
f13: 0x4046d3b6 flt: 3.10667181
f14: 0x70a3d70a flt: 4.05648183e+29 dbl: -67.659999999999997
f15: 0xc050ea3d flt: -3.26429677
f16: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f17: 0x7ff80000 flt: nan(0x780000)
f18: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f19: 0x7ff80000 flt: nan(0x780000)
f20: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f21: 0x7ff80000 flt: nan(0x780000)
f22: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f23: 0x7ff80000 flt: nan(0x780000)
f24: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f25: 0x7ff80000 flt: nan(0x780000)
f26: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f27: 0x7ff80000 flt: nan(0x780000)
f28: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f29: 0x7ff80000 flt: nan(0x780000)
f30: 0x7ff80000 flt: nan(0x780000) dbl: nan(0x800007ff80000)
f31: 0x7ff80000 flt: nan(0x780000)
fcsr fir restart
00800000 00f30000 00000000
(gdb)
I think we need to get the ability to align output (previously achieved
with formatters, i.e. "%-17.9g" and "%-24.17g") restored; frankly I find
the new format unreadable, which makes me consider it unacceptable.
Let me know if I could help with either issue anyhow.
NB while striving for FP arithmetic to match target hardware accurately
we'll have to handle the two opposite qNaN vs sNaN encodings that the MIPS
architecture now supports, i.e. the original IEEE 754-1985 format with the
quiet bit being 0 and the recent IEEE 754-2008 format with the quiet bit
being 1. This obviously has to be taken into account in calculations,
according to the hardware mode being active (or the relevant ELF file
header flag in the binary chosen to debug if not connected to any target),
and I do hope MPFR can be switched at the run time to do the right thing.
This can of course be the next step, once we've settled on the base
implementation -- and I'll take care of the MIPS architecture bits.
Maciej